28th TONMEISTERTAGUNG – VDT INTERNATIONAL CONVENTION, November 2014
3D speaker management systems – Mixer integration concepts
Etienne Corteel1, Peter Glaettli2, Raphael Foulon1, Reda Frauly1,
Ingo Hahn2, Roger Heiniger2, Renato S. Pellegrini1
1
2
Sonic Emotion Labs, France, Email: etienne.corteel@sonicemotion.com
Studer Professional Audio GmbH, Switzerland, Email: Peter.Glaettli@harman.com
Abstract
3D speaker management systems are often only used when creating special spatial effects is required in a show.
However, a 3D speaker management system can replace a traditional speaker management system and a delay
matrix. Such a 3D system must support main sound reinforcement setups with high power loudspeakers allowing for
homogeneous sound level distribution. Spatial effects must be rendered for the entire audience providing consistent
auditory and visual impressions and an enhanced intelligibility of the sound stage.
What is the difference in the setup process between these two system types? Which usability concepts are needed on
a mixing desk in order to control such 3D speaker management systems? For which situations is a joystick the best
usability choice and which other physical interfaces can be used for controlling the positioning of the 3D sources?
How can the source selection process made easy and how are the sources displayed to the user?
We are presenting concepts here and a practical implementation for the efficient use of 3D speaker management for
everyday production situations combining spatial sound rendering based on Wave Field Synthesis and a mixing
console. The system is capable of spatial sound reinforcement in the entire audience using a comparable number of
loudspeakers to standard installations. We present tools and installation procedures that illustrate the simplicity of the
installation procedure. Furthermore, the integration of the controls in the mixing desk enables optimum control of the
spatial rendering by the sound engineer during the performance in a consistent environment.
1. Introduction
Nowadays, 3D sound systems used in theatres and concert
halls are usually fully detached from the main sound
reinforcement system. We are presenting here a concept for
a fully integrated solution that brings 3D to sound
reinforcement for improved segregation of sound sources on
stage and coherent audio and visual impressions for the
entire audience that can be directly controlled from a mixing
console.
The integration alleviates the use of multiple components
such as a delay matrix, 32 bands equalizer and loudspeaker
management systems. All these functions are available in the
Sonic Wave 1 processor with additional dynamic 3D control
of multiple sound sources based on Wave Field Synthesis.
Positioning information of sound sources can be directly
controlled within the mixing desk as a standard function for
best efficiency during setup and performance.
In this paper, we first introduce Wave Field Synthesis for
practical spatial sound reinforcement using a similar number
of loudspeakers compared to today’s installations. We then
present concrete applications and references. Tools for
configuration of the system and visualization of 3D sources
are presented. Finally, we describe in details the integration
of controls for 3D speaker management within Studer Vista
mixing desk.
2. Wave Field Synthesis for Spatial
Sound Reinforcement
2.1. Spatial sound reinforcement
Spatial sound reinforcement aims at providing the audience a
spatial impression during live performances. Spatial sound
reinforcement puts however heavy requirements on the
spatial sound rendering technique being used, imposing the
spatial rendering to be effective throughout the audience,
hence in a very large listening area.
Different sources need to be reinforced (either live or prerecorded) that may modify the spatial rendering
requirements but also the type of interaction for the sound
engineer:
•
•
Live sources with a corresponding real source that
should be properly localized on stage:
o
individual sources like voice, guitar (single
channel, can move);
o
extended objects like choir, piano, drums,
orchestra section (most often static but can
move during performance);
sounds with no visual reference onstage :
o
effects or recorded sounds that may be
located on stage or all around the room;
o
reverb (multichannel static object).
28th TONMEISTERTAGUNG – VDT INTERNATIONAL CONVENTION, November 2014
2.2. Benefits of Wave Field Synthesis
Wave Field Synthesis (WFS) is a sound field rendering
technique that allows for spatial sound reproduction over a
large listening area [1][2][3]. WFS aims at recreating the
sound field of acoustic sources using an ensemble of
loudspeakers positioned at the boundaries of the listening
area. The target sound sources can be freely positioned in
angle but also distance. WFS therefore allows for the proper
reproduction of the parallax effect (angular localization
changes based on listener position in space).
The parallax effect is illustrated on figure 1 where two
sources located on stage are observed/heard from three
different listening positions. It can be seen that none of the
listeners observe/hear the sources from the same direction.
of the auditory system [6] improving the ability of listeners
to segregate the different sounds being reinforcement.
Spatial reproduction also reduces the masking of concurrent
sound sources being played concurrently. This so-called
“spatial unmasking” is known to improve the intelligibility
of sound stage [7] and is closely related to the well-known
cocktail party effect [8].
The naturally enhanced intelligibility provided by spatial
sound reinforcement limits the use of EQs and compression
during the performance in order to unmask concurrent
sources (e.g. voice vs. keyboard), which contributes to the
naturalness of the sound reinforcement.
2.3. Practical installations with optimum sound
level coverage
For practical sound reinforcement applications, WFS
however suffers from three major limitations:
Figure 1: parallax effect for sound reinforcement
The “intensity” of the parallax effect depends on the distance
of the source from the audience: the closer the more angular
localization differences are observed throughout the
audience. The distance of the source can therefore be
considered as a homogeneity parameter for the audience,
more distance meaning more homogeneous impressions for
the entire audience (more similar angular localization) [4].
For sound reinforcement, the reproduction of the parallax
effect (distance control of sources on stage) provides
coherent visual and auditory impressions for any seat in the
audience. It therefore guaranties a consistent localization of
sound sources on stage throughout the audience. This is a
unique property of WFS that no other reproduction
technique can offer within an extended listening area.
The distance of the sources may however be exaggerated to
guarantee a more homogeneous level coverage along the
width of the hall while limiting the angular perception
difference. This may result in localization errors for listening
positions opposite to the source position. However, the
audience may not notice this mismatch for small angular
differences between the auditory and the visual localization.
The ventriloquism effect allows a disparity of 5 to 15
degrees depending on the type of sound without breaking the
audiovisual integration [5].
Beyond localization accuracy, WFS offers natural spatial
auditory cues that are known to favour the streaming process
1.
The large number of loudspeakers required:
traditional WFS imposes a loudspeaker spacing that
should not exceed 15 to 20 cm requiring tens to
hundreds of loudspeakers even for frontal only
installations.
2.
The use of compact low efficiency/dynamic range
loudspeakers: the close spacing imposes the use of
compact
loudspeakers
that
offer
limited
performance and do not match today’s level
requirements for sound reinforcement.
3.
A necessary trade off between vertical localization
accuracy and level coverage: only 1 horizontal
array is used for sound reproduction that can be
practically arranged above the stage, providing
erroneous localization for front seats, or on/below
stage thus ending up in poor level coverage for rear
seats due to masking and natural attenuation of the
loudspeaker array.
The WFS algorithm integrated in the Sonic Wave I
processor alleviates these limitations using a small number
of high power/efficiency loudspeakers that are often
distributed over two levels (above and on/below the stage)
for homogeneous level range distribution and spatial sound
reproduction keeping all benefits of WFS. Corteel et al.
introduced this concept of “multiple level WFS” for sound
reinforcement in [8].
The loudspeakers located on or below stage can be
considered as a “spatial” front fill array. The loudspeaker
ensemble typically spans the entire stage width using 8 to 16
wide dispersion speakers (5 to 8 inches coaxial or
asymmetric directivity speakers with wide horizontal
dispersion).
3 to 7 loudspeakers are commonly used above the stage
targeting homogeneous level coverage further back in the
room. Loudspeakers should either be point source
loudspeakers with asymmetric directivity (wide horizontal
dispersion, directive behaviour along the vertical dimension)
or line arrays with preferably wide horizontal dispersion.
28th TONMEISTERTAGUNG – VDT INTERNATIONAL CONVENTION, November 2014
Typical LCR configurations are applicable with two large
line arrays on either side of the stage and a central cluster.
The stage system can be complemented with surround
speakers for positioning sources outside of the stage
opening.
Sonic Emotion’s algorithm is also offering full 3D WFS
rendering by adding loudspeakers in height. The algorithm
only requires a reduced number of loudspeakers, similar to
the number of surround speakers. The algorithm shows very
accurate height localization accuracy within an extended
listening area [11].
Figure 2: Classique au vert/Paris Jazz Festival layout.
3. Practical examples
3.2. Theatre
In this section we present practical usage of WFS for spatial
sound reinforcement in various use cases.
The sound reinforcement system that has been installed at
the Piccolo Teatro in Milan for the 12-scene play "The
Panic" written by Argentinian playwright Rafael
Spregelburd and directed by Luca Ronconi in 2013 [9]. In
2014, a second production directed by Luca Ronconi (La
Celestina written by Fernando de Rojas) was created at the
Piccolo Teatro using the Sonic Wave 1 counting up to 50
shows in two years.
3.1. Front of house
The Sonic Wave1 processor has been used for two major
summer festivals at the Parc Floral in Paris: the Paris Jazz
Festival and Classique au Vert. More than 400000 spectators
were present over 2 years (2013 and 2014).
The open-air concert stage is located below a tent with a
1400 seats fan shaped area. A standing area behind the seats
can host an additional audience up to 2000 extra people.
The system is composed of a standard LCR line array
configuration. 10 8 inch coaxial speakers are used on stage
as a WFS front fill to complement the system. The sound
engineer Jacques Laville (Urbs Audio1) installed and tuned
the installation.
More than 60 shows used the system ranging from solo
piano to classical orchestra and jazz ensembles. No
adjustments in terms of speaker positions were necessary.
Sounds were routed from the mixing desk generating up to
24 streams to the Sonic Wave1 (direct out for solo
instruments, groups for extended sound sources or reverb).
Jacques Laville acted as a regisseur and sound engineer for
about half of the shows. Jacques Laville had prepared the
console layout and introduced the system to visiting sound
engineers. They adjusted jointly the source positions on
stage to fit with the stage layout. The visiting sound engineer
could then work as usual, monitoring all other parameters
(levels, Eq, compression) directly from the console, leaving
out panning information.
Ronconi's request to sound designer Hubert Westkemper
was to enable all members of the audience seated anywhere
in the venue to hear each amplified voice of the 16-member
cast coming from the position of the actor in question even
when moving around the stage.
The theatre has more than 500 seats in the stalls and about
400 in the balcony, while the dimension of the stage is about
22 m x 19 m. The speaker setup consisted in 2 horizontal
loudspeaker arrays at different height levels to guarantee an
optimum level coverage over the entire audience. The lower
loudspeaker array was composed of 24 5 inches speakers
(wide high frequency dispersion), with an about 70 cm
spacing, positioned along the stage front and covering the
front half of the stalls. The front speakers were masked with
black acoustical fabric, to ensure no visual localization bias.
The array was completed by 2 sub-woofers positioned on the
left and right side of the proscenium.
The sound engineers appreciated the transparency of the
system and the additional intelligibility that resulted from the
spatial rendering. The sound engineers but also the
production team, the visiting artists and the audience
perceived the additional comfort. A testimonial video was
created during the 2013 edition of the Classique au Vert
festival (https://www.youtube.com/watch?v=b5ZVJZtckNk).
Figure 3: Piccolo Teatro, Milano, The Panic 2013
1
http://urbsaudio.fr/
28th TONMEISTERTAGUNG – VDT INTERNATIONAL CONVENTION, November 2014
The upper array was installed above the proscenium at a
height of about 6.5 m. It was composed of 8 pairs of
loudspeakers with a spacing of about 2 m and a minimum
distance from the audience of 12 m. These loudspeakers,
already part of the sound system of the theatre itself, are
more directive (70°×40°). The upper speakers of the 8 pairs
were aiming to the balcony, while the lower ones covered
the second half of the stalls, partly under the balcony for
optimum SPL coverage.
All 18 actors were equipped with wireless microphones and
had their movements tracked on stage. All actors had one,
some even two, Ubisense tags, detected by 5 sensors
positioned around the stage, hence to locate the actors'
position. The actor’s position was automatically transmitted
to the Sonic Wave 1 processor that updated the source
position accordingly. 6 additional tracks were used for pre
recorded sounds that were located at fixed positions on
stage.
3.3. Concert hall
Radio France (Paris, France) has equipped the Studio 105
that hosts a large number of live events with an audience.
The existing LCR line array installation plus sub has been
complemented with 28 additional loudspeakers (8 inches
coaxial): 10 for WFS front fill and 18 surrounding. A
temporary installation was using different speakers from
September 2013 until September 2014 where the installation
was completed.
Due to the diversity of the stations of Radio France, the
studio is hosting a large variety of events: radio shows,
concerts (pop, rock, jazz, world music, classical music,
contemporary music, electro, …) or occasionally
conferences (Forum International du Son Multicanal 20122014 for which the WFS system has become an essential
piece).
The installation combines the possibility of frontal sound
reinforcement and surrounding effects/reverb. The
surrounding speakers can be used to enhance the room
perception by using multichannel reverberation units or by
simply duplicating stereophonic reverb over multiple virtual
sources located all around the audience. Effects can be in the
form of a multichannel content (5.1 or 7.1) that can be
reproduced using “virtual loudspeakers” that mimic a
multichannel setup or individual objects that can be freely
positioned in real time.
The studio is used by up to 15 sound engineers from Radio
France who are being trained. Over the past year, the full
WFS system has been used in average twice a month for live
shows with an audience. The studio is now regularly
requested specifically for the use of the WFS system.
Figure 4: Studio 105, Maison de la Radio, Paris.
4. Control options and interfaces
Two software interfaces have been designed to control
the Sonic Wave1 processor. The Wave Designer, is used for
configuration of the system (loudspeaker positioning, output
routing, system tuning). The Wave Performer is used during
the performance for 3D sound positioning and visualization.
4.1. OSC communication
The Sonic Wave1 processor can be controlled using the OSC
protocol [10], a widely used communication standard
designed for musical applications. OSC messages are based
on the UDP protocol, and consist of a string header followed
by additional arguments. A set of messages is specified2 to
offer control of the audio sources 3D position, level,
equalization, naming and routing. In particular, “pair
sources” messages enable the user to efficiently control
stereo pairs as a single entity, by specifying it’s angular
width and position.
The use of such a simple protocol and specification enable
users to control the Wave processor with OSC-based
programs that let the user create custom interfaces, or
patches. Such interfaces can be run either on mobile devices
using applications such as Liine’s Lemur3 and Hexler’s
TouchOSC4, or on desktop devices using standalone
software such as Max/MSP5 and PureData6.
4.2. Wave Designer, configuration and tuning
interface
The Wave Designer presents itself as a computer-aided
design interface used to create loudspeaker setups. The
Wave Designer has two modes:
1.
An offline mode for loudspeaker positioning and
output routing
2
available on request at pro@sonicemotion.com
3
http://liine.net/en/products/lemur
4
http://hexler.net/software/touchosc
5
http://cycling74.com/products/max
6
http://puredata.info
28th TONMEISTERTAGUNG – VDT INTERNATIONAL CONVENTION, November 2014
2.
4.2.1.
An online mode for system tuning.
Offline configuration
In the offline mode, the user creates a loudspeaker
configuration by positioning speakers in 3D space. The
speakers can be either created individually, or by generating
linear or circular arrays of loudspeakers.
The creation of front of house systems is therefore be made
very easy by declaring one line for the WFS front fill
speakers indicating the stage width and the number of
speakers. The upper loudspeaker array is similarly created as
well as eventual additional surrounding speakers.
This basic setup being created, positions of individual
speakers can be adjusted to match with realistic installation
constraints (stairway, doors, other equipment). Sonic
Emotion’s WFS algorithm does not require regular
loudspeaker spacing to operate in best conditions.
processor workload. Prior to the upload, the Designer
performs a sanity check of the setup to validate the
configuration and indicate to the user the potential errors.
4.2.2.
Online tuning
The Wave Designer turns into online mode for further setup
& tuning.
A simple test interface offers an automatic or manual tool for
testing individual loudspeakers and routing by selecting a
given input of the system and routing it directly to a given
speaker/output.
The online mode of the Wave Designer contains a complete
parametric equalizer for each loudspeaker with high and low
shelving filters, high and low pass filters with up to 48 dB
slope, and 8 peak filters. Furthermore, each parametric
equalizer can be adjusted according to 3 different modes:
•
individually per loudspeaker;
•
semi-globally by subsystem (LCR, WFS front-fill,
surround, …);
•
globally for all loudspeakers.
This way, an ensemble of speakers can be tuned very
efficiently, offering a comprehensive grouping of speakers
of similar type (line arrays or point source) and located in
similar acoustical conditions (simply put on stage, hang in
height, attached to walls, …).
Figure 5: Wave Designer interface in offline mode, Studio
105 Radio France concert hall setup
The equalization can be set manually or by loading
equalization presets (from loudspeaker manufacturers or
stored by the user).
The speakers can be organized in groups, that we call
subsystems. For instance, in Radio France concert hall, four
subsystems are declared. Each subsystem represents a
specific section of the setup: the WFS front fill, the
surrounding speakers, the LCR line array and the sub.
The subsystems are assigned a spatial rendering algorithm,
which can be either “WFS main” or “WFS support”. All
subsystems in “WFS main” mode are then being grouped to
form the principal WFS rendering system that can either be
fully peripheric or restricted to the front. Subsystems in
“WFS support” mode are considered as complementary
loudspeaker arrays that are used to bring energy at further
locations in the room. Such a “support array” can be the
LCR line array configuration and additional under balcony
loudspeaker arrays. An adapted Wave Field Synthesis
algorithm drives these “support arrays” for creating a
consistent spatial sound rendering in support to the main
WFS array.
Loudspeakers are assigned to physical outputs in a simple
routing matrix. For multi-way loudspeakers, one physical
input is assigned for each frequency range (low, mid, high).
Once the configuration is created, it is uploaded to the
processor that creates the right number of speakers and
outputs and calculates WFS filtering for reducing online pre-
Figure 6: Wave Designer interface in online mode
Sonic Emotion’s WFS algorithms expect that each speaker
within the main WFS setup and within each support
subsystem deliver the same acoustical power.
Level adjustment can be realized for each speaker by
measuring SPL at 1 m distance with a regular sound level
meter. However, using identical speakers within each WFS
main system/support subsystems is most often a good
enough approximation.
The Online Mode of the Wave Designer further offers
loudspeaker management for multi-way speakers allowing to
control crossover frequencies; level, delay phase inversion
and a simple compressor for each physical output of the
28th TONMEISTERTAGUNG – VDT INTERNATIONAL CONVENTION, November 2014
system. These settings can also be stored with the
loudspeaker equalization preset.
An additional panel offers simple level and delay
adjustments for each subsystem allowing aligning/balancing
the main WFS array and the support subsystems.
The complete configuration (loudspeaker positioning, output
routing, equalization, loudspeaker management, level/delay
alignment) is stored as a preset for archiving the
configuration and loading it at a later time or on a different
machine.
4.3. Wave Performer, performance visualizer
interface
The Wave Performer serves several purposes: it offers 2D
and 3D displays of the scene, and enables sound engineers to
move audio sources during a performance.
In order to clarify the sound scene, sources can be grouped,
hidden, assigned names and are presented as a list on the
right side of the interface.
Figure 7: Wave Performer source positioning interface
The Wave Performer also includes a master equalization,
which can be used to adapt the overall sound colour of the
installation. The equalization is fully independent from the
one offered in the online mode of the Wave Designer and
keeps the configuration unaltered.
This is useful to sound engineers who tour with the same
equipment. Moreover, when various sound engineers use the
system, for instance, in festivals, the master equalization is
used to provide each sound engineer with a specific sound
coloration. It can fully replace a typical 32 band graphical
equalizer of a classical setup.
Additionally, each source can be directly routed to one or
several loudspeakers and/or subsystems, in a routing
module. This allows applying a specific audio rendering to
each source, in order to adjust the system to various sound
reinforcement situations. Finally, the Performer provides a
source level mixer with a display of the input levels.
Each performance configuration is storable as an xml preset
file, so that user-specific configurations can easily be
recalled.
5. Mixing desk integration
As outlined above a 3D speaker management system is an
independent (sub-) system within a live performance setup.
Especially the configuration tasks are normally carried out
completely with the user interfaces of the speaker
management system. In case of a 3D speaker management
system though a tight integration with the mixing desk
significantly simplifies the operation of the complete system
during the performance. In the following the essential
control functions and integration concepts for an effective
operation during the live performance are presented.
5.1. Retrieve source signals; naming of sources
In order to do a proper channel setup on the mixing desk the
list of all (signal) sources needs to be retrieved from the 3D
speaker management system. Since the mix is normally
organized on the mixing desk the naming (labelling) of these
‘shared’ sources needs to be pushed back from the mixer to
the 3D speaker management system. The 3D speaker
management system imports and displays these labels in the
graphical user interface (GUI) of the performance visualizer
tool. By these means the sources can be recognized with the
same name/label in both systems.
Selecting a source in the performance visualizer tool is not
quite straightforward. The sources are normally presented as
small circles (or similar marks) with limited naming
information. Moreover, several sources are often positioned
at almost the same place in the sound space, which makes it
very difficult to distinguish them in the GUI. Zooming
functions or browsing through a list can help but it is also
rather cumbersome in order to quickly select a source in the
performance visualizer.
The natural place for the source selection is on the mixing
desk where a configurable and a cue recallable strip setup
organizes the channels, which are important for a specific
scene.
In addition a lot of times the sound engineer does not only
want to change the spatial position of a source but also
controls other channel processing parameters. The mixing
desk therefore distributes the channel selection information
to the 3D speaker management system and this source
selection information is then displayed within the
performance visualizer tool.
5.2. Positioning of sources
As outlined above there are different source types that have
different interaction requirements in order to do the spatial
positioning: the individual sources, the extended objects and
the effects.
28th TONMEISTERTAGUNG – VDT INTERNATIONAL CONVENTION, November 2014
The simplest use case is the positioning of the individual
sources (single channel). For this source type we propose to
patch the direct out of the input channel to the corresponding
source in the 3D speaker management system. The standard
panning control of this input channel can be reused to
calculate the x/y (or azimuth/distance) coordinates within the
3D sound space. A new ‘3D positioning mode’ is added to
the panning control and when activated the mixing desk
sends the x/y coordinates to the 3D speaker management
system. By these means the user has to control only one
panning model if the same source signal is also used in a mix
within the mixer.
The joystick, which controls the same panning parameters, is
better suited for the control of rather fast and dynamic spatial
effects or for the outline positioning of a source. Due to the
limited resolution and control area of the built in joysticks of
mixing desks an exact positioning of a source is
cumbersome though.
Besides these individual sources the 3D speaker
management system should also be capable of controlling
extended objects or pre-panned/stereo sources. An example
of such a pre-panned source would be a choir. The different
voices are routed to a stereo group and each voice has its
stereo-panned send signals. The picture below shows the
setup of these voices within a choir.
Sopran
Tenor
Tenor
Bass
Left
Right
Figure 8: Example of a pre-panning for a choir.
The extended object (or stereo group) positioning within the
3D speaker system not only needs the x/y positioning
information but also the width and the rotation parameters as
depicted below.
CLOUD
Wi
dth
Center Position (X/Y/Z)
STAGE
Rotation
FRONT
Figure 7: 5.1 panning control in 3D mode.
Using rotaries allows positioning the sources very accurately
in the sound space – the sound engineer can change the
parameters with a precise tactile control, gets the acoustic
feedback from the sound system and can see the source
positioning in the performance visualizer tool. As soon as a
panning rotary control is touched the mixing desk
automatically sends a corresponding ‘source selected’
message to the 3D speaker management system and the user
can see, which source position is changed. The rotaries are
also ideal for controlling rather precise but slow movements
of these sources in a scene.
Figure 9: Width and rotation parameters; example for a
movement of the choir in a scene
If the choir would walk from the back to the front of a stage
in a certain scene all of these parameters would have to be
controlled accordingly. Depending on the speed of such a
movement the control through rotaries is still possible. In a
programmed show the crossfade function of the mixing desk
automation system can be used defining only the start and
end positions and the crossfade time.
If the z-axis is added to the sound space the control of the
source positioning gets even more complicated. For the
28th TONMEISTERTAGUNG – VDT INTERNATIONAL CONVENTION, November 2014
rather static source positioning the rotaries can still be used.
But (faster, dynamic) movements in the x/y/z space are not
easy to control with the rotaries only. One approach is to
combine the joystick (for the x/y positioning) with a rotary
control (z-axis positioning). Another option could be the use
of systems, which recognize hand gestures. But these
systems lack of a tactile feedback for the user. This is the
root cause for most operation errors in such systems. Ultra
haptic devices are one solution to this problem. These
devices create a haptic feedback for the user with the aid of
ultra sonic waves.
(MADI) is appropriate. An AoIP interface might allow a
more dynamic selection of input signals in a later stage.
A simple control protocol should be designed so that both
systems are not coupled to tightly with each other and so that
the line up (debugging) is easy. OSC is a standard control
protocol, which is widely used for third party control in the
professional audio domain and offers all the means in order
to implement the functions of the integration interface
outlined above.
Figure 10: Example of an ultra haptic device in laboratory
application.
First research projects investigate this technology – but it is
too early to evaluate whether these devices are the right
choice for controlling sound source movements in the x/y/z
sound space. The resolution of these devices is promised to
be similar to today’s GUI controls (button, slider, rotary,
fader).
5.3. Snapshot and automation
A speaker management system should feature an
independent snapshot module (create, update, recall, delete
snapshot), which can be accessed through a public control
interface. Such an interface is perfect for changing
configuration or setup parameters of the speaker
management system during scene changes. But the
automation of all relevant parameters for the source
positioning is simpler within the mixing desk. The source
positioning parameters become part of the native snapshot
and automation module of the mixing desk and inherit
functions like isolation (masking) and parameter crossfade.
By these means the interface between the speaker
management system and the mixer is simpler (lightweight)
and the overall snapshot recall performance is more precise.
Editing the snapshot data offline is also simplified by this
approach since the user has to work only in one system.
5.4. Interfaces
The integration of a 3D speaker management system to a
mixing desk requires an audio and a control interface. Since
the signal sources for the speaker management system all
stem form the mixer a simple point-to-point connection
Figure 11: system diagram with 3D speaker management
and mixing system.
6. Conclusion
An effective integration of a 3D speaker management system
into the workflow of the mixing desk was presented. It is
essential for a simple operation of the overall system during
a live performance. A comprehensive interface displays the
overall sound stage showing either all or a limited subset of
sources in a 2D and a 3D view.
The most important aspect is the management of the source
positioning parameters using standard controls of the
console surface.
The configuration and tuning of the system is realized very
efficiently by easily providing loudspeaker positioning
information. The system does not require aligning gains and
delays of a large number of individual loudspeakers but only
loudspeaker ensembles (so-called sub-systems). Such
simplification results in faster setup times and more
flexibility during the use of the system. Configurations can
be prepared offline and recalled or modified during
28th TONMEISTERTAGUNG – VDT INTERNATIONAL CONVENTION, November 2014
installation. This includes loudspeaker positioning/routing
but also system tuning information (eq, loudspeaker
management, system alignment).
The 3D speaker management system also offers the
possibility of spatial sound reinforcement for enhanced
intelligibility of the sound stage and reduced masking of
competing sources, ensuring an improved overall sound
quality.
7. References
[1] A. J. Berkhout, D. de Vries, and P. Vogel, “Acoustic
control by wave field synthesis,” Journal of the
Acoustical Society of America, Vol. 93, No. 11, pp.
2764-2778 (1993).
[2] S. Spors, R. Rabenstein, and J. Ahrens, “The theory of
wave field synthesis revisited,” presented at the 124th
Convention of the Audio Engineering Society (May
2008), convention paper 7358
[3] E. W. Start, “Direct Sound Enhancement by Wave
Field Synthesis,” PhD thesis, TU Delft, Delft, The
Netherlands (1997)
[4] Noguès M., Corteel E., Warusfel O. “Monitoring
Distance Effect with Wave Field Synthesis”, Proc. of
the 6th Int. Conference on Digital Audio Effects
(DAFX-03), London, UK, September 8-11, 2003
[5] Perrott “Auditory and visual localization: Two
modalities, one world,” 12th International Conference
of the Audio Engineering Society, 1993.
[6] A. S. Bregman, “Auditory Scene Analysis: The
perceptual organization of sound” (MIT Press) (1990).
[7] N. I. Durlach and H. S. Colburn, “Handbook of
perception, volume IV: Hearing, chapter Binaural
phenomena,” 365– 466 (New York: Academic) (1978).
[8] Corteel E., Damien A., Ihssen C. “Spatial sound
reinforcement using Wave Field Synthesis. A case
study at the Institut du Monde Arabe,” 27th
TonmeisterTagung - VDT International Convention,
November, 2012
[9] Corteel E., Westkemper H., Ihssen C., Nguyen KV. :
“Spatial Sound Reinforcement Using Wave Field
Synthesis,” 134th convention of the Audio Engineering
Society, May 2013.
[10] Wright, M., Freed, A.: “Open Sound Control: a new
protocol for communicating with sound synthesizers,”
Proc. of the 1997 Int. Computer Music Conf., pp. 101–
4. Thessaloniki, Hellas: ICMA, (1997).
[11] Rohr L., Corteel E., Nguyen KV, Lissek H. “Vertical
localization performance in a practical 3D WFS
formulation,” Journal of the Audio Engineering
Society, Vol. 61, No. 12, pp. 1001-1014 2013.